johns said:
I can't see any difference at all. I don't think an LCD
monitor is scanned line by line. If it was, I could wave
my fingers in front of my eyes, and the scan would
show up as stop motion of my fingers like it does
with a crt. I've read articles where the author talked
about the scan rate of an LCD monitor, but I'm
officially declaring that to be bullshit ! I think the
screen dots are digitally addressed just like ram,
which would make the update super fast, and
produce no strobe action at all.
johns
There is a PDF icon on this page. There is a datasheet for a bare
panel (the part with the LCD plus backlight but no inverters) in
the PDF.
http://www.semirim.com/main/products/lcd/sharp/index.cgi?l=c&c=693
http://datasheet.semirim.com/ds/vie...ys/tft/pcmonitor/tec_specs_lq201u1lw01.pdf.gz
A later member from the same panel family here.
http://document.sharpsma.com/files/LQ201U1LW21_SP_101105.pdf
If you look at section 7 "Timing characteristics of input signals",
the 1200 vertical lines of the display are scanned in a
period from 12.1 to 16.7 milliseconds. 16.7 milliseconds
corresponds to selecting 60Hz. The 12.1 number is as fast as
the panel will go (where the vertical sync period before the
1200 lines is at its minimum period). The 12.1msec number is
82.6Hz. Which means the panel could work at 60Hz, 72Hz, 75Hz,
but no higher.
Before reading that, I always thought the refresh of the LCD
was at 60Hz, no matter what. But the above docs do suggest the
panel is actually updated faster, when you select a higher
refresh rate.
If it not possible to say anything about internal panel operation,
because in fact the control method may bear no relationship
to that interface spec.
This article shows the LCD elements are multiplexed. There are
row and column wires. Sequentially scanning rows and columns allows
assigning a voltage to a picture element.
http://en.wikipedia.org/wiki/TFT_LCD
More info here showing the multiplexing. This implies scanning
of a matrix of pixels, by applying voltage to the rows and columns.
Each pixel stores the voltage pulse applied, until the next
scanning pass. The polarity of the voltage must alternate,
so that no net DC voltage (bias) is applied to the crystals.
(It means, for a given picture element, if you looked at the
voltage across the element, it is an AC waveform. The amplitude
of the waveform, determines whether the element is relatively
opened or closed.) As long as the storage capacitor does its job,
there would be no scanning artifact to be seen by moving your
fingers in front of the display.
http://www.avdeals.com/classroom/what_is_tft_lcd.htm
I vaguely remember mention, that panels can be broken down into
a number of rectangular areas. The scanning process can operate on
the rectangular areas independently, so that the dwell time per pixel
can be longer. (A long dwell time may be required, if the TFT
transistor switches slowly.) I don't know if I'd be able to find
a datasheet for a control ASIC for inside the actual panel, to find
out if that is what they are doing or not. Controlling 1200 and 1600 lines
discretely, does not sound possible for the low price of LCD monitors,
and perhaps the scanning process is actually built into the panel
glass itself.
OK, there are some screen drivers depicted here.
http://www.samsung.com/PressCenter/PressRelease/PressRelease.asp?seq=20040729_0000066447
The driver ICs listed here, don't seem to have enough signals to drive the
whole array with just one chip. Getting a datasheet requires
registration (oh,well).
http://www.samsung.com/Products/Semiconductor/common/product_list.aspx?family_cd=DDI0201
Pretty complex tech for a $300 monitor.
Paul